High-Efficiency Thermoelectrics with Functionalized Graphene - Nano

Subscriber access provided by RUTGERS UNIVERSITY

Communication

High-Efficiency Thermoelectrics with Functionalized Graphene Jeong Yun Kim, and Jeffrey C. Grossman Nano Lett., Just Accepted Manuscript • DOI: 10.1021/nl504257q • Publication Date (Web): 06 Apr 2015 Downloaded from http://pubs.acs.org on April 11, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Nano Letters is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

High-Efficiency Thermoelectrics with Functionalized Graphene Jeong Yun Kim and Jeffrey C. Grossman* Department of Materials Science and Engineering, Massachusetts Institute of Technology, Cambridge, MA 02139, USA * Author to whom the correspondence should be addressed: [email protected]

Abstract Graphene superlattices made with chemical functionalization offer the possibility of tuning both the thermal and electronic properties via nano-patterning of the graphene surface. Using classical and quantum mechanical calculations, we predict that suitable chemical functionalization of graphene can introduce peaks in the density of states at the band edge that result in a large enhancement in the Seebeck coefficient, leading to an increase in the room-temperature power factor of a factor of two compared to pristine graphene, despite the degraded electrical conductivity. Furthermore, the presence of patterns on graphene reduces the thermal conductivity, which when taken together, leads to an increase in the figure of merit for functionalized graphene by up to 2 orders of magnitude over that of pristine graphene, reaching its maximum ZT ~ 3 at

ACS Paragon Plus Environment

1

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 24

room temperature according to our calculations. These results suggest that appropriate chemical functionalization could lead to efficient graphene-based thermoelectric materials.

Keywords: graphene, chemical functionalization, transport, thermoelectrics Recent progress in enhancing the figure of merit for thermoelectric materials has relied on a range of innovative strategies, including reduction of the dimensionality,1-3 use of complex bulk materials,4, 5 or introduction of impurity states in the bulk phase.6, 7 These advances, based on modifications to three-dimensional (3D) crystalline materials, have been enabled by a deep understanding of the key thermoelectric (TE) properties such as thermal and electronic transport, and the impact of structural and chemical changes on these properties, in turn providing new design strategies for high-efficiency TE materials. In contrast to the case of 3D materials, twodimensional (2D) materials such as graphene, single-layer boron nitride, or molybdenum disulfide, have received little attention regarding their potential for thermoelectric applications. Given that 2D materials are “all-surface,” any modification to their environment or chemical functionalization will be expected to have enormous impact on properties, an appealing attribute when faced with such constrained optimization problems as in the case of thermoelectrics. Chemical functionalization, in particular, plays a key role for both thermal and electronic transport in 2D materials since either all or nearly all atoms in the system can be functionalized, leading to large modifications of charge carriers8 and phonon modes;9 this is in contrast to the 3D case where surface chemical functionalization has a much smaller impact on transport. While such potentially controllable transport in 2D is appealing for thermoelectric applications, a unified understanding of the key TE properties in 2D materials is needed in order to assess the


2

Page 3 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

potential of 2D TE and to provide design strategies for increasing the efficiency of this new class of TE materials. Recent work has shown that the most familiar 2D carbon material, namely graphene, exhibits remarkable electronic transport properties, with a record carrier mobility of ~200,000 cm2 V-1 s-1 reported for a suspended single layer10 as well as a moderate Seebeck coefficient (S) of ~80 µV K-1.11, 12 Graphene is well known to possess an ambipolar nature, such that the sign of S can be controlled by changing the gate bias instead of doping.13 Experimentally measured values of the thermoelectric figure of merit (ZT) for pristine graphene at room temperature are in the range of 0.1 − 0.01,11, 12, 14, 15 roughly two orders of magnitude lower than the most efficient 3D thermoelectric materials such as Bi2Te3 alloys.16 Such low ZT values for graphene are to be expected due to its extremely high thermal conductivity (κ), on the order of 2000−5000 W m -1K -1 .15, 17, 18 As mentioned above, graphene can be modified controllably with chemical

functionalization, which in turn modulates its electronic and thermal transport properties; a range of methods have been employed to fabricate graphene heterostructures, e.g., hydrogen functionalization,19 introduction of in-sheet regions of hexagonal boron nitride (h-BN),20 or the attachment of polymers using patterning processes such as lithography.21 These controllable patterning techniques facilitate routes to the formation of a range of confined graphene patterns, including dots, lines and more complex structures within a single sheet of graphene. In this work, we consider the prospect of patterned graphene nano-roads22 as an efficient thermoelectric

material.

The

patterning

introduces

quantum

confinement

in

pure

(unfunctionalized) graphene by reducing its dimensionality from 2D to quasi-1D, with the periodic functionalization lines forming arrays of parallel graphene regions. Confinement in the material leads to an improvement in the power factor by increasing the density of states at the


3

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 24

Fermi level, as demonstrated in recent studies1-3 for low-dimensional thermoelectric materials. Here, ab initio electronic structure calculations and the Boltzmann transport approach are used to determine the role of such functionalization on the electrical transport properties, including the electrical conductivity (σ) and Seebeck coefficient (S), which combined with our computed thermal conductivities enables us to predict the full ZT in these materials. Our results show that ZT can reach values as high as 3 at room temperature, and demonstrate the potential for controllably tuning the properties of 2D materials for thermoelectric applications. Methodology. Room-temperature transport calculations of patterned graphene nano-roads are carried out using electronic structure calculations and the Boltzmann transport approach. Firstprinciples calculations are performed within the framework of density functional theory (DFT) as implemented in the plane-wave basis VASP code.23 The projected augmented wave pseudopotentials24 with the PW9125 generalized gradient approximation of Perdew and Wang is used for treatment of the exchange-correlation energy. Geometry optimizations are performed using the conjugate gradient scheme until force components on every atom are less than 0.01 eV Å-1. The fully relaxed structures are used to compute the electronic band structure. Transport coefficients are calculated by employing semi-classical Boltzmann theory within the constant relaxation time approximation, as implemented in the BOLTZTRAP code.26 The electrical conductivity (σ) and Seebeck coefficient (S) are calculated by defining the following function

L(α ) = e2 ∑ ∫ n

dk ⎛ ∂ f (ε nk ) ⎞ − τ n (ε nk )vnk vnk (ε nk − µ )α 3⎜ ⎟ 4π ⎝ ∂ε nk ⎠

(1)

where e is the electronic charge, τ n the energy-dependent relaxation time, vnk = (1 / )∇ kε nk the group velocity in the nth band at k , ε nk the energy eigenvalues obtained from DFT, µ the


4

Page 5 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

chemical potential, and f (ε nk ) the Fermi-Dirac function at a given temperature T. Here, we are interested in transport along the pattern direction (y), which is obtained from σ = L(0) yy and −1 (1) S = −(1 / eT )(L(0) yy ) L yy . The converged values of σ and S are obtained using a dense k-point

mesh of 121 points along the y direction. All simulations are carried out at 300 K and a carrier concentration range of 1011−1013 cm-2. In general, the relaxation time (τ) is taken to be energy-independent, and obtained by fitting experimental values for pristine bulk materials even in the cases with structural and chemical modifications. However, in principle τ is a complex function of the electron energy, temperature, and atomic structure. In this respect, using a τ value obtained from pristine graphene is not a reasonable approximation for use in patterned graphene nano-roads, where chemical modification in those samples plays an important role in opening the energy gap, which indirectly relates to the relaxation time. Since the goal of our work is to determine the quantitative effect of functionalization on charge transport, we calculate the relaxation time of functionalized graphene samples with Boltzmann transport theory for a reliable quantitative approximation. Here, we consider only the acoustic phonon scattering, and the relaxation time can be expressed as 1 2π k BTDac2 = ∑δ (ε k − ε k' ) τ (ε k ) A ρ m v 2ph k'

(2)

where k B is the Boltzmann constant, Dac is the deformation-potential coupling constant, ρ m the graphene mass density, v ph the graphene sound velocity, and A the area of the sample. The term

∑δ (ε

k

− ε k' ) A yields the density of states of functionalized graphene. This form highlights the

k'


5

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 24

proportionality of the scattering rate to the density of states of the 2D graphene sheet (see Supporting Information). For calculations of the thermal transport, we employ equilibrium molecular dynamics (MD) simulations as discussed in ref. 9, which shows a large κ reduction in functionalized graphene with the use of 2D periodic patterns. The MD simulations describe well the scattering effects of functionalization on thermal transport in graphene, especially for short-wavelength phonons. However, given the current limits of system size and timescales, our simulations do not capture the contributions of long-wavelength phonons, which are important for thermal transport in graphene, resulting in lower κ values in our simulations by an order of magnitude than measured values. In order to predict ZT, we modify our computed κ values by adding the long-wavelength phonon contributions from the Klemens model,27, 28 allowing one to obtain κ values for larger graphene samples (see Supporting Information).9

Figure 1. (a) Schematic of 2D thermoelectric based on patterned graphene nano-roads with an armchair-type boundary. Yellow shaded area indicates the patterned region. (b) We consider Hfunctionalized graphene (HG) for 6 Å pattern width (top) and C5-functionalized graphene (C5G) (bottom). C and H atoms are represented by gray and yellow spheres, respectively. Pink contour plot represents charge density of the conduction band in each sample. Green and blue spheres


6

Page 7 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

indicate the C sites where C5 chains are attached on the up and down side of graphene, respectively. (c) Density of states (DOS) per unit cell of graphene, HG and C5G. Atomic Structure. Figure 1a shows a schematic of a 2D thermoelectric device based on graphene nano-roads with armchair-type interfaces formed via partial functionalization. This type of superlattice is modeled with a repeated structure of pristine graphene (sp2 network) and graphene with functionalization (sp2 + sp3 network) within the same sheet, where the former exhibits semi-metallic behavior, and the latter shows insulating or semiconducting behavior depending on the functional groups. The electronic structure of functionalized graphene is strongly influenced by the functionalized configurations on the graphene sheet, determined by the shape of the functional groups.

For example, H-functionalization generates full

functionalization (perfect sp3 network) showing insulating behavior, whereas hydrocarbon chains generate partial functionalization of the graphene sheet due to the steric repulsion between the chains, exhibiting a graphene-like electronic structure with a distorted cone-shaped band. In this work, we consider two different functional groups (H and pentane C5H12, referred to as C5) to investigate how chemical functionalization influences the thermoelectric properties of a patterned graphene superlattice. H-Functionalized Graphene. We first compute the electronic structure of partially Hfunctionalized graphene (HG). The width (W) of the pattern in our simulations ranges from 6−23 Å and can be represented by the number of pure carbon dimers (NC) along the pattern direction in the unit cell as shown in Figure 2a. While graphene is a zero-gap semiconductor with a 2D cone-shape linear energy dispersion, all HG samples show semiconducting behavior with a finite energy gap due to confinement effects of the graphene parts of the structure. As in the case of armchair graphene-based nano-ribbons,29 the gap can be categorized into three different families


7

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 24

depending on N C : Eg ( N C = 3p + 2 ) 0.15 eV, S2σ decreases as Eg increases, indicating that the wider H-pattern generates superior charge transport in these graphene superlattices. The reduction in S2σ in this regime is mainly due to more severe scattering for narrower H-patterns. On the other hand, S2σ decreases as Eg decreases for Eg < 0.15 eV, and this behavior can be explained by the S2 term in the power factor, which is significantly reduced with H-pattern width, due to weaker confinement effects. Thus, the HG sample with the NC = 8


11

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 24

pattern exhibits the best performance for TE applications with S2σ ~ 0.93 W m-1K-2, which is about 1.6 times larger than pristine graphene. We note that this large enhancement in power factor arises from considerably enhanced S as well as a comparatively small reduction in σ in spite of strong scattering due to the H-pattern.

Figure 3. (a) Configuration of C5-functionalized graphene (C5G) with the structure of C5 (pentane: C5H12). C5 chains are arranged in a two-dimensional hexagonal close-packed lattice on the graphene sheet. (b) Relaxation time for C5G as a function of the carrier energy. C5G_all represents C5G sample with full coverage. (c) Deformation potential constant ( Dac ) and corresponding power factor of C5G samples, where * represents calculated Dac values for C5G samples. (d) Power factor of fully functionalized C5G as a function of spatial direction. The black dashed line indicates the average value of the power factor. The background represents the energy band contour of constant energy, showing a distorted cone-shape dispersion.


12

Page 13 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

C5-Functionalized Graphene. We next consider the case of functionalizing graphene with pentane chains (C5), which was shown previously9 to further reduce κ in graphene superlattices beyond simple hydrogen functionalization due to a clamping effect imposed by the steric hindrance between chains (see Supporting Information). In contrast to the one-dimensional dispersion of HG cases, C5-functionalized graphene (C5G) exhibits a 2D distorted cone-shape band, resulting in an anisotropic σ (Figure 3d). This difference is caused by the fact that C5 patterns cannot generate confinement effects in graphene due to their inability to pack them closely enough to functionalize every carbon atom. As a result, the DOS of a C5G sample is similar to that of pristine graphene (Figure 1c), showing no peak at the band edge, and our computed S values are similar in all directions to that of pristine graphene. Among the several C5G samples we considered with different pattern widths, the case of fully C5-functionalized graphene (i.e., width=0, so no “nano-roads”) exhibits the largest S2σ, with an average value close to the value of pristine graphene (0.486 W m-1K-2). The peak in S2σ ~ 0.88 W m-1K-2 arises at θ = 30°, where the group velocity along this direction is largest due to the distorted band dispersion (Figure 3d). It should be noted that in contrast to HG samples, charge carriers in C5G samples are not confined in an un-functionalized region but rather delocalized over the entire 2D graphene sheet as shown in Figure 1b. Accordingly, we need to determine an accurate value of Dac for C5G samples, which considers the electron-phonon coupling for this specific functionalization. To this end, we compute Dac using Bardeen and Shockley’s theory,35 and the calculated Dac and corresponding power factors are shown in Figure 3c (see Supporting Information). Reduction in Dac gives rise to an increase in τ for the C5G samples, which is inversely proportional to the

square of Dac , as shown in Figure 3b. It is interesting to note that C5-functionalization leads to


13

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 24

the suppression of electron-phonon coupling in the graphene sheet, implying diminished charge scattering. Thus, the values of S2σ for C5G samples are close to the value of graphene, showing huge potential for TE applications when combined with the significantly reduced κ for this type of functionalization, which was shown previously.9 H-functionalization gives rise to confinement of the graphene region, where confined charge carriers induce an enhanced power factor, while the confined phonon modes in this case have little effect on thermal transport along the pattern direction, resulting in 3× lower thermal conductivity than that of pristine graphene.9 In contrast, C5-functionalization greatly suppresses κ in graphene superlattices from the steric repulsion of the C5 chains,9 although as shown above no power factor improvement is observed for this case. Thus, each functionalization type alone would not be sufficient to make graphene-based TE practical. However, combining both functionalizations into the same sample could achieve the best of both worlds: namely, controllable charge transport with H-functionalization and tunable thermal transport with C5functionalization. In this combined case, a well-defined boundary is formed between the sp2 graphene and sp3 domains, giving rise to the same types of electronic confinement effects as in the HG cases, and therefore the same energy dispersions, resulting in significant enhancements in S2σ. We can also achieve further suppression of κ compared with either the HG or C5G samples due to a combination of boundary scattering and clamping effects.


14

Page 15 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 4. Thermoelectric properties of patterned graphene nano-roads with different functional groups. Calculated thermal conductivity (green), power factor (orange) and ZT (red) of patterned graphene as a function of functional groups are shown for NC = 8 pattern width, which exhibits the most enhanced charge transport. These are computed values along the pattern direction. Using the correction for κ with Klemens model, we show our calculated thermoelectric properties of patterned graphene nano-roads as a function of functional groups in Figure 4 for the NC = 8 pattern width. We predict maximum value of ZT ~ 0.09 for pristine graphene at room temperature, in good agreement with experimental measurements of intrinsic thermoelectric properties for graphene monolayer without any extrinsic scattering sources (see Supporting Information).33 By combining the κ reduction and S2σ enhancements in each sample, we obtain significantly increased ZT values for all functionalized graphene samples compared to the pristine graphene case. Among these cases, a factor of ~ 30 enhancement in ZT can be achieved in the C5+H sample, with corresponding ZT values as large as 3 at room temperature. This large enhancement in ZT is due to a combination of a large reduction in κ and relatively small


15

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 24

enhancement in S2σ by 40 % from pristine graphene. Interestingly, the fully C5-functionalized graphene case also yields a figure of merit well above 2 at room temperature, even without periodic patterning on the graphene sheet, suggesting that chemical functionalization by itself offers further opportunities for graphene to become fascinating TE material without complex nano-patterning. The results suggest that chemical functionalization could be an important route to designing high-efficiency graphene-based TE materials, by allowing for independent control of charge transport and thermal transport. In this work, we investigated the thermoelectric properties of patterned graphene nano-roads functionalized with H and C5 chains using both classical and quantum mechanical calculations. Periodically patterned sp3-hybridized carbon regions on graphene were shown to introduce DOS peaks at the band edge, which lead to substantially enhanced S and S2σ compared to that of pristine graphene at carrier concentrations corresponding to the band edge. These results also demonstrate that the thermal transport in such materials can be tuned by functional groups separately from the power factor, which simply depends on the pattern width. We note that unlike the case of standard, 3-dimensional TE materials, for the 2D cases considered here it is not difficult to achieve reduced κ and enhanced S2σ simultaneously, resulting in broad tunability of thermoelectric properties for this class of materials. Finally, although fabrication of functionalized graphene with 100 % H coverage or regular C5 chain ordering may be challenging, we note that incomplete H-functionalization in the patterned region has a negligible influence on ZT values of HG and (C5+H)-G samples since charge transport in these samples is mainly determined by the condition of the pristine graphene region. Furthermore, our calculations of a fully functionalized C5G sample with random (as opposed to regular) C5 chain ordering still shows high ZT values at room temperature, as the decrease in thermal conductivity


16

Page 17 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

roughly tracks the decrease in the power factor, indicating that perfectly controlled ordering is not crucial for the potential use of functionalized graphene in TE applications.

Supporting Information Available Additional simulation details on thermal conductivity calculations (molecular dynamics simulations and Klemens model), relaxation time calculations, and thermoelectric coefficient as a function of carrier concentration for pristine graphene, HG, and C5G samples. This material is available free of charge via the Internet at http://pubs.acs.org.

References 1.

Venkatasubramanian, R.; Siivola, E.; Colpitts, T.; O'Quinn, B. Nature 2001, 413, 597-

602. 2.

Dresselhaus, M. S.; Chen, G.; Tang, M. Y.; Yang, R. G.; Lee, H.; Wang, D. Z.; Ren, Z.

F.; Fleurial, J. P.; Gogna, P. Adv. Mater. 2007, 19, 1043-1053. 3.

Lee, J. H.; Galli, G. A.; Grossman, J. C. Nano Lett. 2008, 8, 3750-3754.

4.

Snyder, G. J.; Toberer, E. S. Nat. Mater. 2008, 7, 105-114.

5.

Uher, C., Skutterudites: Prospective novel thermoelectrics. Academic Press ed.;

Semiconductors and Semimetals: San Diego, 2001; Vol. 69, p 139-253. 6.

Heremans, J. P.; Jovovic, V.; Toberer, E. S.; Saramat, A.; Kurosaki, K.; Charoenphakdee,

A.; Yamanaka, S.; Snyder, G. J. Science 2008, 321, 554-557. 7.

Lee, J. H.; Wu, J. Q.; Grossman, J. C. Phys. Rev. Lett. 2010, 104, 016602.

8.

Georgakilas, V.; Otyepka, M.; Bourlinos, A. B.; Chandra, V.; Kim, N.; Kemp, K. C.;

Hobza, P.; Zboril, R.; Kim, K. S. Chem. Rev. (Washington, DC, U. S.) 2012, 112, 6156-6214.


17

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 24

9.

Kim, J. Y.; Lee, J. H.; Grossman, J. C. ACS Nano 2012, 6, 9050-9057.

10.

Bolotin, K. I.; Sikes, K. J.; Jiang, Z.; Klima, M.; Fudenberg, G.; Hone, J.; Kim, P.;

Stormer, H. L. Solid State Commun. 2008, 146, 351-355. 11.

Seol, J. H.; Jo, I.; Moore, A. L.; Lindsay, L.; Aitken, Z. H.; Pettes, M. T.; Li, X. S.; Yao,

Z.; Huang, R.; Broido, D.; Mingo, N.; Ruoff, R. S.; Shi, L. Science 2010, 328, 213-216. 12.

Zuev, Y. M.; Chang, W.; Kim, P. Phys. Rev. Lett. 2009, 102, 096807.

13.

Geim, A. K.; Novoselov, K. S. Nat. Mater. 2007, 6, 183-191.

14.

Bolotin, K. I.; Sikes, K. J.; Hone, J.; Stormer, H. L.; Kim, P. Phys. Rev. Lett. 2008, 101,

096802. 15.

Balandin, A. A. Nat. Mater. 2011, 10, 569-581.

16.

Poudel, B.; Hao, Q.; Ma, Y.; Lan, Y. C.; Minnich, A.; Yu, B.; Yan, X. A.; Wang, D. Z.;

Muto, A.; Vashaee, D.; Chen, X. Y.; Liu, J. M.; Dresselhaus, M. S.; Chen, G.; Ren, Z. F. Science 2008, 320, 634-638. 17.

Balandin, A. A.; Ghosh, S.; Bao, W. Z.; Calizo, I.; Teweldebrhan, D.; Miao, F.; Lau, C.

N. Nano Lett. 2008, 8, 902-907. 18.

Vincent, E. D.; Ashkan Behnam; Hiram J. Conley; Kirill I. Bolotin; Pop, a. E. Nano Lett.

2013, 13, 4581–4586. 19.

Sun, Z. Z.; Pint, C. L.; Marcano, D. C.; Zhang, C. G.; Yao, J.; Ruan, G. D.; Yan, Z.; Zhu,

Y.; Hauge, R. H.; Tour, J. M. Nat. Commun. 2011, 2, 559. 20.

Liu, Z.; Ma, L. L.; Shi, G.; Zhou, W.; Gong, Y. J.; Lei, S. D.; Yang, X. B.; Zhang, J. N.;

Yu, J. J.; Hackenberg, K. P.; Babakhani, A.; Idrobo, J. C.; Vajtai, R.; Lou, J.; Ajayan, P. M. Nat. Nanotech. 2013, 8, 119-124.


18

Page 19 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

21.

Wang, Q. H.; Jin, Z.; Kim, K. K.; Hilmer, A. J.; Paulus, G. L. C.; Shih, C. J.; Ham, M.

H.; Sanchez-Yamagishi, J. D.; Watanabe, K.; Taniguchi, T.; Kong, J.; Jarillo-Herrero, P.; Strano, M. S. Nat. Chem. 2012, 4, 724-732. 22.

Singh, A. K.; Yakobson, B. I. Nano Lett. 2009, 9, 1540-1543.

23.

Kresse, G.; Furthmuller, J. Phys. Rev. B 1996, 54, 11169-11186.

24.

Kresse, G.; Joubert, D. Phys. Rev. B 1999, 59, 1758-1775.

25.

Perdew, J. P.; Wang, Y. Phys. Rev. B 1992, 45, 13244-13249.

26.

Madsen, G. K. H.; Singh, D. J. Comput. Phys. Commun. 2006, 175, 67-71.

27.

Klemens, P. G. J. Wide Bandgap Mater. 2000, 7, 332-339.

28.

Chen, S. S.; Wu, Q. Z.; Mishra, C.; Kang, J. Y.; Zhang, H. J.; Cho, K. J.; Cai, W. W.;

Balandin, A. A.; Ruoff, R. S. Nat. Mater. 2012, 11, 203-207. 29.

Son, Y. W.; Cohen, M. L.; Louie, S. G. Phys. Rev. Lett. 2006, 97, 216803.

30.

Lee, J. H.; Grossman, J. C. Phys. Rev. B 2011, 84, 113413.

31.

Mahan, G. D.; Sofo, J. O. Proc. Natl Acad. Sci. USA 1996, 93, 7436-7439.

32.

Pennington, G.; Goldsman, N.; Akturk, A.; Wickenden, A. E. Appl. Phys. Lett. 2007, 90.

33.

Chen, J. H.; Jang, C.; Xiao, S. D.; Ishigami, M.; Fuhrer, M. S. Nat. Nanotech. 2008, 3,

206-209. 34.

Fang, T.; Konar, A.; Xing, H.; Jena, D. Phys. Rev. B 2008, 78.

35.

Bardeen, J.; Shockley, W. Physical Review 1950, 80, 72-80.


19

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

66x26mm (300 x 300 DPI)


Page 20 of 24

Page 21 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

232x354mm (300 x 300 DPI)


Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

116x89mm (300 x 300 DPI)


Page 22 of 24

Page 23 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

55x37mm (300 x 300 DPI)


Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

44x26mm (300 x 300 DPI)


Page 24 of 24

High-Efficiency Thermoelectrics with Functionalized Graphene - Nano

Recommend Documents